Epidemiology of abscess and cellulitis among United States emergency departments from 2016 to 2023

Abstract

Skin and soft tissue infections are a common emergency department (ED) presentation. Data suggested a rapid rise in ED presentations from 1993 to 2005 when community-acquired methicillin-resistant Staphylococcus aureus (CA-MRSA) first appeared.1 However, there are limited recent data on incidence and admission rates since that time. Moreover, with changes in resistance rates, advances in antibiotic options, increasing antibiotic stewardship, and evolving literature surrounding the role of antibiotics in abscess, there is a need to better understand the distribution of antibiotic use.2 This study reports the incidence, admission rates, and antibiotic distribution among a nationwide cohort of ED patients with abscess and cellulitis over an 8-year period. We performed a cross-sectional study using Epic System Corporation's Cosmos research platform, an application that aggregates electronic health record data submitted voluntarily by health systems for research purposes. Cosmos data are representative of national Census data (https://cosmos.epic.com/). At the time of this study, the Cosmos data set included 238 million unique patients and 31,500 hospitals and clinics. Patients with records at more than one institution were deduplicated and anonymized centrally by Epic. We queried Cosmos using ICD-10 codes for cellulitis and abscess (Supplemental File 1—Appendix S1) from January 1, 2016, to December 31, 2023. This time period was selected based upon limitations of Cosmos data prior to 2016 and the launch of ICD-10 coding requirements in October 2015. This study adheres to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.3 The Rush University Medical Center Institutional Review Board deemed this study exempt. Out of 197,327,418 ED encounters over the 8-year period, 1.04% were due to abscess and 0.36% to cellulitis. Of these, 21.36% of abscesses and 13.08% of cellulitis were admitted. From 2016 to 2023, there was an overall decline in ED presentations for abscesses, ranging from 1.28% of all visits in 2016 to 0.91% in 2023 (Figure S1, Table S1). In contrast, presentations for cellulitis remained relatively stable over time. However, among those presenting to the ED, admission rates increased for both abscess and cellulitis over the 8-year period (Figure S2, Tables S2 and S3). Among ED patients with abscess, the most common discharge antibiotic was trimethoprim-sulfamethoxazole (Figure 1A). While trimethoprim-sulfamethoxazole and clindamycin use declined over time, there was a notable increase in doxycycline and amoxicillin-clavulanate. The most common antibiotic for patients admitted with an abscess was vancomycin, which continued to increase over time (Figure 1B). Among ED patients with cellulitis, the most common discharge antibiotic was cephalexin (Figure 1C). Similar to patients discharged with an abscess, trimethoprim-sulfamethoxazole and clindamycin use declined over time, while there was a notable increase in doxycycline and amoxicillin-clavulanate. Among admitted patients with cellulitis, the most common antibiotic was also vancomycin, while linezolid use nearly doubled during that time (Figure 1D). While previous research reported a substantial increase in ED visits during the first decade after CA-MRSA first appeared,1 our study found that the proportion of ED visits for these has begun to decline. This may be due to increased access to primary care or comfort with managing these conditions in a primary care setting, earlier recognition and initiation of appropriate antibiotics in the primary care setting, or reduced severity of infection over time. Alternatively, this may reflect overall rising rates of ED visits over time for alternate reasons, leading to a proportionally lower rise in visits for abscess.4 Moreover, the shift in visit types due to COVID-19, as well as the recent expansion of telemedicine programs, may have further influenced this.5, 6 Future research should follow these rates over time to better understand the impact of these recent factors on the trend in visit proportions. Additionally, we identified several trends in antibiotic prescribing over time. Clindamycin use has declined, which may reflect rising resistance rates or increased awareness of side effects.7 This has been associated with commensurate rises in agents such as doxycycline for oral regimens and either vancomycin or linezolid for intravenous regiments. The rise in intravenous vancomycin and linezolid use for cellulitis is concerning as many cases should be treated with narrower spectrum agents unless there is a high suspicion for a severe, necrotizing infection or CA-MRSA is suspected.8 Overuse of newer agents such as linezolid when not clinically indicated may increase antimicrobial resistance rates. Therefore, appropriate antimicrobial stewardship is critical. Finally, the overall high rate of antibiotics for abscesses may also suggest reduced comfort with incision and drainage alone. This study was limited by the data in the electronic health records and may have missed some cases that were coded incorrectly. Moreover, by relying upon ICD-10 codes, some diseases process may have been missed or incorrectly coded. Additionally, by using large-scale deidentified data, it is possible that some antibiotics may have been administered for other concomitant infections not coded. Despite a large number of total encounters, we were limited to those within the Epic system. The overall total number of encounters also rose over time, which may have influenced the proportions. Finally, we are not able to analyze differences in the etiologic agent (e.g., MRSA vs. non-MRSA), abscess severity, or the influence of sites outside the network on these proportions and future research should be conducted to further understand the influence of these factors. In summary, this study provides data on the current epidemiology of skin and soft tissue infections as well as the trends in antibiotic usage over an 8-year period. These findings provide important insights to inform health policy and antibiotic stewardship for skin and soft tissue infections. The authors declare no conflicts of interest. Appendix S1: Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Topics & Concepts